machine rigidity testing linear rails and ball screws

I'm trying to rebuild an old Fanuc tape drill.

The machining performance was terrible so I assumed it was the spindle taper which was in really bad condition.
The spindle has been reground and the runout on the tool taper (DA180) is less than .0005”, but it still can’t side mill to an acceptable finish unless it is going very slowly.

I can see there are tool deflection issues. It leaves a vertical line on the side if you stop mid cut and leave the spindle running. The tool deflection could be caused by lots of things and I’m trying to narrow it down.

I need to determine if the tool deflection is in the spindle bearings, the linear ways, or backlash in the ball screws. It is probably in all three, but what is the worst.

Has anyone got some test procedures I could follow to narrow down the cause?

What model? I have worked on B thru F. I ask so I know what you are dealing with. The B thru D models had funky Z axis ways with box X and Y.

I like to go through the machine using an indicator and try to isolate ways from nuts from thrust bearings, and write it all down. Push and pull using a relatively even amount of pressure, say 75 lbs or so. The machine must be powered up but there may still be some servo drift when testing the nuts and thrust bearings. Test the head for sideways movement in both x and y axis. How square is the head to the table?

You are not using DA holders to mill with correct?

I thought I would look at what you have. Fanuc Drill Mate - YouTube

You may ask this question in the CNC forums also.

I was thinking it was a bigger machine at first and wrote some last night. Now I edited it.

The linear ways: Run it all the way up and feel and visualize the LW's, look for grooving. Also put a mag base on column and .0001" indicator on headstock and check for lost motion (described below)"shake" by pushing and pulling on the head. You may have to use a small piece of wood or carefully slide in a brass bar between the LW and headstock on both sides of the head. Check at the top and at the bottom of head, record your readings. Then move the head down at the bottom of the LW's and check it again for shake. I would bet it shakes all over the place. Do you have a manual?


I would first get some Prussian blue or what ever paste bluing you have down there. If you don't know go to an auto parts store and ask them what they sell for testing valve seat in heads. Then blue up your best tool holder. Spread an even layer but thin so it's transparent or you can see thru it. slide it in the spindle taper tight and look at the holder and see where the bluing rubs off the holder. also use a mirror and bright flashlight and look up in there.

If that seems loose remove the 2 square keys and blue it up again. You will want minimum of 80% contact. If that looks good then Put a shaft in a 1" holder say it sticks out 8" then mount a mag base on the head stock NOT THE TABLE and then put a .0001" indicator on the spindle and push and pull (have to lean on it...if your a little guy put a long Allen bolt with washer and jam nut in tee slot and use as 2 x 4 about 24" long and lever /pry it ..not super hard..but you should be able to see something easy if it's loose.... look for "lost motion" that's if you push / pull it and release it record the location the spindle will bend a little but should go back to the zero. If it bends more then .001" and if it doesn't go back to zero and has "lost motion your bearings are loose. Also at the auto store buy a auto stethoscope and listen to the bearing. They make the most noise when the coast after shutting it off.

Try that and see what happens.. Member RC99 and McTools ( I thinks are rebuilders down there, they ight be able to help if they are close. Bed time for me. Good luck

That's a new one Richard! This is an older one with the funky Z axis. CNC Retrofit Pratt & Whitney Tapemate C (Fanuc Drill Mate) Making First Part - YouTube

landslide said:

....... It leaves a vertical line on the side if you stop mid cut and leave the spindle running.

Click to expand...

Don't all machines do this even in the multi-million dollar range?
Maybe add servo tuning to your possible problem list.
Bob

I think I'd run from a machine like that. It's basically a glorified Burgmaster drill. Milling is never going to be it's strong suit.

The model F I ran worked fine for milling as long as you didn't get too rambunctious. It was still nice and tight when it was 20 years old which was when I left. It would use a 3/8" mill just fine, for what it was. Even the C or D models would mill surprisingly well.

I see these with the P+W logo. Did P+W build them or buy and relabel? Who invented the concept?

They were Fanuc with a P&W label. If you needed parts or tech help you called Fanuc, at least in 2000 or so.

I was thinking that most machines leaves a line when you stop / dwell it when you stop midpoint. trying to fix something that isn't an issue. Again you should ask this in the CNC Milling forum. More people with these small machine watch and read that forum and can be of more help because they see these issues. We talk more about scraping machines here. Rich

any pics of the issue? A small cutter is going to have deflection.

thanks everyone.

mine is a model T 1986 vintage.

I'm sure it has limitations but I have developed a soft spot for it and I'm not giving up on it yet.

after viewing some of those videos I'm sure the milling performance can be improved. My machine is shaking and complaining on similar cuts.

The spindle taper contact is perfect now it is rebuilt. The spindle has angular contacts at the taper and only a deep grove at the drive end. Bearings were only P5 so I’m not sure if this is going to be a source of deflection.

You can see the large riser plate on the vid of the more modern drill tap machine above. I made my own and it is no-where near as beefy as that one. I know it has some deflection and I’m going to remove it for further tests.

Also from the sound of the vids, my rails and ball screws are stuffed. They sound so loud and gritty in comparison. Probably expected since this machine came from a dirty old foundry.

From the sounds of Richard kings suggestions I’m performing similar tests but didn’t know about the lost motion issue.
I’ll see if I can get some pictures of the finish. If I go as slow as the vids above finish is quite good side milling but gets worse the faster I go (as expected).

Yikes. I have a Burgmaster that's a lot beefier than that thing. I think you will have an uphill battle to make it mill anything productively.

I have never seen a burgmaster but from a google search it is similar to the very early tape drills. The model T i have the predecessor to the robodrill.
Sure it is better at drilling, but I’m not wanting to push it. I Just want better finish.

You may want to google this question. I recall a few years ago on another board how some people were doing all sorts of things to their Chinese made crap machines to make them more rigid. This board usually stays away from such instruction as it's dedicated to professional machine tools. Rich

One thing to try is a torque test. First, move x and y to the center of travel. Put a vise on the table at the center. Put an indicator on the edge of the table about 18" from center. Grab the square drive of a torque wrench in the center of the vise and apply 100 ft pounds of torque. Observe the deflection on the DI. This will test both the x and y ways for play. On my Kitamura, I get a few tenths. I am guessing you will get much more.

I have been a bit distracted to do any real testing. just a quick push on the spindle with a 6-8" 1"dia tool. there was about 0.0015" in the bearings and 0.0005" dead movment as it didn't return. I'm going to talk to the spindle re-grinder and check the pre-load again.

here are some pics of the finish. The worst is an 8mm 4 flute cutter (it was in a DA180) driven with the hand wheel not a set G01 feed. the lower pass was G01 feed stop and restart. The other pic is sidemilling with a sandvik r390 in a weldon holder.

Your pics clearly show the value of using a real tool holder instead of a DA180. When troubleshooting always start with the simplest / cheapest possible fix. Tool holders are much cheaper / easier than linear rails / spindle bearings. Your tape drill machine will never do a great job milling but picture #2 shows definite promise!

After some more testing I can say the finish in the photo wasn't the DA180 fault. It was caused by inconsistant feed with the hand wheel.
I have some ER32 collets on the way so I'll see what difference they make, but when programed the marks are like the lower cut.

End milling cutters are called that for a reason because they flex in a cut when side milling, try a larger diameter cutter using climb feed as this will reduce the power needed to cut, with ball screws climb feed is usable,increasing speed will give a better finish.